Miles Brundage announced the AI Verification and Evaluation Research Institute (AVERI), a nonprofit to advocate for independent, third‑party audits of frontier AI models and for audit standards across the industry. Brundage, a former OpenAI policy researcher, argued companies currently “write their own rules” for safety testing and that independent verification should become standard practice. The launch coincides with broader research from labs such as Anthropic showing AI’s uneven effects on jobs, reinforcing calls that governance and assessment frameworks keep pace with rapid model deployment.